Update app.py

app.py

@@ -153,7 +153,7 @@ def tokenizer(text: str):
     """
     phonemes_string, tokens = g2p(text)
     phonemes = [ph for ph in phonemes_string]
-    print(phonemes_string)
+    print(text + "     " + phonemes_string)
     tokens = [phoneme_vocab[phoneme] for phoneme in phonemes if phoneme in phoneme_vocab]
     return tokens
 
@@ -197,7 +197,7 @@ def tts_streaming(text: str, voice: str = "af_heart", speed: float = 1.0, format
             if i == 0:
                 tokens_to_send = [0] + chunk_tokens + [0]
             else:
-                tokens_to_send = [prev_last_token] + [16] + chunk_tokens + [0]
+                tokens_to_send = [prev_last_token] + [16] + [0] + chunk_tokens + [0]
             #     token_to_send = [0] + chunk_tokens + [0]
 
             # Save the last token of this chunk for the next iteration.
@@ -205,7 +205,7 @@ def tts_streaming(text: str, voice: str = "af_heart", speed: float = 1.0, format
 
             # Prepare the model input (a batch of one sequence).
             final_token = [tokens_to_send]
-            print(final_token)
+            # print(final_token)
 
             # Use the number of tokens to select the appropriate style vector.
             style_index = len(chunk_tokens) + 2